Frank, S. A. 2012. Wright's adaptive landscape versus Fisher's fundamental theorem. Pages 41-57 in The Adaptive Landscape in Evolutionary Biology. E. Svensson and R. Calsbeek, eds. Oxford University Press.

Two giants of evolutionary theory, Sewall Wright and R. A. Fisher, fought bitterly for over thirty years. The Wright-Fisher controversy forms a cornerstone of the history and philosophy of biology. I argue that the standard interpretations of the Wright-Fisher controversy do not accurately represent the ideas and arguments of these two key historical figures. The usual account contrasts the major slogans attached to each name: Wright's adaptive landscape and shifting balance theory of evolution versus Fisher's fundamental theorem of natural selection. These alternative theories are in fact incommensurable. Wright's theory is a detailed dynamical model of evolutionary change in actual populations. Fisher's theory is an abstract invariance and conservation law that, like all physical laws, captures essential features of a system but does not account for all aspects of dynamics in real examples. This key contrast between embodied theories of real cases and abstract laws is missing from prior analyses of Wright versus Fisher. They never argued about this contrast. Instead, the issue at stake in their arguments concerned the actual dynamics of real populations. Both agreed that fluctuations of nonadditive (epistatic) gene combinations play a central role in evolution. Wright emphasized stochastic fluctuations of gene combinations in small, isolated populations. By contrast, Fisher believed that fluctuating selection in large populations was the main cause of fluctuation in nonadditive gene combinations. Close reading shows that widely cited views attributed to Fisher mostly come from what Wright said about Fisher, whereas Fisher's own writings clearly do not support such views.